Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > Manufacturing/Packaging

Litho options falling behind process needs

Posted: 16 Jul 2008 ?? ?Print Version ?Bookmark and Share

Keywords:high-index lithography? EUV? nanoimprint? immersion lithography?

Who killed high-index lithography, and was politics the cause? Those remain pressing questions for chipmakers looking for a route to next-generation manufacturing in a shifting lithography landscape.

Cymer Inc. has reported a breakthrough in extreme UV (EUV) power sources that could burnish EUV's chances as the next-generation lithography (NGL) technology of choice. Separately, Nikon Corp. has dropped its high-index 193nm lithography equipment program, further eroding support for that rival technology.

These events imply that the industry is narrowing the NGL options for chip production at the 32nm node and beyond. Double patterning, EUV, maskless and nanoimprint are still on the table. But maskless and nanoimprint lithography are underfunded and not ready for prime time. That leaves EUV as the strong front runner, despite delays, soaring costs and continued uncertainty about the technology.

One down
Nikon's scrapping of its high-index program is a signal that high-index lithography is losing steam. Rival ASML Holding NV has been lukewarm about the technology, and many leading-edge chipmakers!reportedly including Intel, Micron and others!have expressed little or no interest in it.

High-index immersion lithography uses a non-water solution to expose a wafer!a scheme that would, in theory, allow the IC industry to extend 193nm optical lithography to the 22nm node and beyond.

Today's 193nm immersion lithography uses plain water to process wafers, thereby boosting the depth of focus in patterning. Immersion has a refractive-index rate of 1.44, which means the technology could hit a wall at the 32nm or 22nm node.

In comparison, high-index uses a complex, non-aqueous fluid to process the wafers in a 193nm scanner, which could boost the refractive index at least to 1.70. In theory, with high-index fluids, 193nm lithography could be extended to the 22nm node and beyond.

But high-index technology could also be a nightmare to develop. It would require new infrastructure, including costly fluids, lens materials and tool platforms. It also requires an ample supply of a rare material called lutetium aluminum garnet (LuAG), which would be used for the lenses in high-index immersion lithography tools.

Schott Lithotec, a division of Schott AG, said it had met its absorption targets for LuAG. That progress moved the company one step closer to meeting its goal of commercially producing 160mm-diameter LuAG for the lithography industry by Q4 09.

But in a paper, Soichi Owa, a fellow at Nikon, said high-index "is falling behind" the International Technology Roadmap for Semiconductors and "is too late" to meet the insertion deadlines of 2008 or 2009. Moreover, leading-edge chipmakers have spent an inordinate amount of time and millions of dollars to get rival EUV technology off the ground. Some IC vendors have put most of their eggs in the EUV basket, and for them, there appears to be no turning back.

Losing options
The industry has long been confronted with the fact that there are few!if any!lithography options for the 32nm node and beyond. That implies a potential disaster in the making for lithography, arguably the most critical part of the semiconductor production process.

"I do not think that it would be right to assign blame," said Klaus Rinnen, an analyst with Gartner Inc. "I think the situation is a combination of events that leaves the industry potentially exposed. The bigger question is: Does the industry understand the problem, and what will it do about it?"

There are no simple answers. At the 45nm node, which is the leading edge in the IC industry today, many chipmakers have begun to process wafers with their initial 193nm immersion lithography scanners. For the 32nm node, it appears that the leading lithography candidate for IC production is 193nm immersion with dreaded!and expensive!double-patterning techniques. Currently, chipmakers are processing wafer in scanners with single-exposure techniques. In double patterning, the wafer is basically exposed twice, thereby creating more cost and complexity for chipmakers.

Beyond 32nm, there are no viable solutions. "The clock is ticking for the industry to find a solution for 22nm," Rinnen, with fellow Gartner analyst Jim Hines, reported in a newsletter. "Urgency is rising to find solutions quickly, because the industry does not have a clear solution to meet the lithography requirements at the 22nm node."

A survey from Sematech indicated the solution for nominal 32nm "half pitch" in 2013 is double patterning using 193nm lithography tools, whether "dry" or water-based immersion types. EUV is still considered the most likely next-generation choice for the 22nm half-pitch node.

Using a 13.5nm wavelength, EUV is a major departure from today's conventional lithographic tools, based on optics and photons. The processing steps take place in a multimirror vacuum chamber. The optical elements are defect-free mirrors, which reflect light by means of interlayer interference.

EUV was originally supposed to process wafers at the 65nm node. But the technology has experienced a series of setbacks, pushing its arrival out to the latter stages of the 22nm node. Some believe EUV has made progress; others say they are convinced the technology will never work.

EUV still suffers from a lack of viable photomasks, photoresists and power sources. Cost is a problem: An EUV tool could run $50 million or more per unit.

ASML and Nikon are working on "preproduction" EUV tools, which are due out in late 2009 or 2010. In May, Cymer provided some reason for hope. With its power source, the company generated 25W of average EUV power continuously for 1.5hrs. Cymer said it is on track to achieve the 100W of average power needed for the first delivery of "production ready" EUV lithography tools in 2009.

Gigaphoton, Philips Extreme UV, Xtreme Technologies and others are working on EUV sources as well. "We have invested a lot over the past years in EUV, with no return on investment and still no assurance that it is the winning technology, despite what some say," Rinnen said. "We need more choices. But those are woefully under-funded. Alternatives to EUV are developed by small firms with shallow pockets and little capital reserves." The analyst was referring to high-index, maskless and nanoimprint lithography

- Mark LaPedus
EE Times

Article Comments - Litho options falling behind process...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top